The purpose of the research is to investigate how natural perceptual and semantic categories are coded, processed, and formed. The major hypothesis is the most natural categories are structured according to a prototype (the best examples) and degrees of perceived distance from the prototype--contrary to the model provided of Part I is to demonstrate that, both in real and natural-like categories, distance from the prototype--and not criterial attributes--predicts diverse dependent variables derived from both cognitive and psycholinguistic research: reaction-time different judgments, category search times, judged probability of items, sentence substitution, and judgments of acceptability of sentences. The research of Part II will employ the learning of categories to determine some of the principles by which prototypes are formed: specifically, the conditions under which attributes are, and are not, coded in terms of their means, and the effects of conjoint frequency of attributes and distance from relevant contrast sets on the characteristics of prototypes developed. Part III explores the nature of stimulus spaces and principles of classification which may initially give rise to naturally organized categories. Using free and constrained classification (sorting) and learning and reaction-time techniques, we will attempt to demonstrate that continuous stimulus spaces become organized around salient stimuli as the centers of categories and that both continuous and discrete stimuli become structured around attribute redundancies (of which we will attempt to show similarity of shape a special case).